Feature 5 March 03 (pp133-144)
27/2/03
8:17 pm
Page 133
Special feature
By Craig Lindsay, Labour Market Division, Office for National Statistics
Key points
A summary of labour market conditions in the twentieth
century.
●
In 1900 the population was
around 38 million and gross domestic
product (GDP) stood at just under
£125 billion at constant 1995 market
prices. By the end of the century, the
population had increased by 50 per
cent to 59 million, and GDP had risen
fivefold to £800 billion.
● The estimated employment rate
for 15 to 64-year-olds in 1902 was
around 69 per cent. Using Labour
Force Survey data the employment
rate in 2000 was 71 per cent.
However, this comparison is affected
by changes in the school leaving age
and retirement age.
● The unemployment rate, as
measured
by
those
claiming
unemployment-related benefit, was
below 5 per cent in 1900. It was at its
highest point in 1932, at 22 per cent
but by the end of the century, it was
below 5 per cent again.
● One major change was the shift in
industrial composition. In the UK,
manufacturing’s share fell from 28 to
14 per cent of employment, and
agriculture’s share from 11 to 2 per
cent.
● At the beginning of the twentieth
century, around five million women
worked, making up 29 per cent of the
total workforce. By 2000, the figure
had risen to 13 million, 46 per cent of
the total workforce.
● The average weekly hours of a
manual worker fell from 53 hours in
1943 to 43.5 in 1987.
● In 1900 trade union membership
represented 11 per cent of those in
employment. It peaked in the late
1970s at 50 per cent but by 2000 was
down to under 29 per cent, its lowest
level for 60 years.
● The level of full-time earnings has
soared from an average £1.40 per
week in 1902 (not adjusted for
inflation) to £350 per week in 1997.
Introduction
THE AIM of this article is to provide an
overview of labour market change
through the twentieth century, and its
links to major events and to social
change more generally. ONS already
produces a range of articles looking at
general labour market conditions: the
monthly Labour Market Assessment
(see pp103-6) looks at the current
labour market situation using the latest
data; the annual ‘State of the Labour
Market’ pieces (of which the first was
published on the National Statistics
website www.statistics.gov.uk last year)
look at change over the course of a year.
By comparison, this article aims to take
a step back and to examine longer-term
trends.
In 1900, Britain was coming to the
end of the Victorian age, a period that
had seen the country at the forefront of
the Industrial Revolution and the
expansion of Empire. Government
intervention was light, and the economy
had developed largely based upon the
free market liberal heritage of Adam
Smith (1723-90), and of Cobden (180465) and Bright (1811-89). As late as
1874, both Disraeli (Prime Minister:
1868-68 and 1874-80) and Gladstone
(Prime Minister: 1868-74, 1880-85,
1886-94) had gone into the general
election promising to repeal income tax
(although during their different times in
office neither did). The nineteenth
century had seen politics dominated by
the Tory and Whig groupings, and then
by their successor parties, the
March 2003
Labour Market trends 133
A century of labour market change: 1900 to 2000
A century of labour market change:
1900 to 2000
Feature 5 March 03 (pp133-144)
A century of labour market change: 1900 to 2000
Special feature
27/2/03
5:09 pm
A century of labour market change: 1900 to 2000
Conservatives
and
Liberals
respectively. However, the beginnings
of change were evident. The Industrial
Revolution, as well as bringing
increased wealth and higher living
standards, had seen the emergence of
the working class and the first disciples
of Marxism.
In 1900 the population was around 38
million1 and gross domestic product
(GDP) stood at just under £125 billion2
at constant 1995 market prices. The
economy was more notably based upon
trade and manufacturing: manufacturing represented 28 per cent of
output; agriculture, forestry and fishing
11 per cent; and services 50 per cent.
Looking at the labour market, the
employment rate was 69 per cent, with
24 per cent in manufacturing and
textiles, and 12 per cent in agriculture.
Unemployment stood at around 3 per
cent. Within this, the workforce was
very much male dominated, with men
representing 70 per cent of the active
population.3
Britain in 2000 was a very different
place. The days of Empire were gone.
The population had increased by 50 per
cent to 59 million;4 by comparison,
GDP had risen fivefold to £800 billion,
at constant 1995 market prices (see
Figure 1). This increase in living
standards was also visible in average
weekly wages, which had risen to over
Figure
1
Page 134
250 times their 1902 level while prices
had risen 67 times.5 Manufacturing’s
importance had declined, representing
only 14 per cent of employment and 22
per cent of output. Similarly, only 2 per
cent of people worked in agriculture,
forestry and fishing, which represented
only 2 per cent of output. By
comparison,
services
represented
around 75 per cent of employment6 and
66 per cent of output. Government
intervention had increased markedly
with, for example, the development of
the Welfare State and the National
Health Service (NHS). Government
expenditure as a proportion of GDP had
increased from 15 per cent to around 40
per cent.7
Looking more generally at the labour
market, the employment rate stood at 71
per cent, with unemployment at 4 per
cent, as measured by the number of
people claiming benefit. Female
participation had increased greatly, with
women representing 45 per cent of the
active working-age population. By
comparison, male participation was
declining.
The twentieth century was a period of
substantial change for the UK. It saw
among other things: two world wars; the
rise and decline of trade unionism; the
Great Depression; unemployment of up
to 22 per cent; and great changes in
social attitudes. There had also been
great political change with the rise of
the Labour Party at the expense of the
Liberals. The political consensus had
shifted over the century: first, to
incorporate greater state provision of
public services; and then more recently
it moved back slightly with
privatisation and the reduction in direct
government intervention in industry. A
similar, if more extreme, shift had been
played out on a global scale with the rise
and fall of communism in Eastern
Europe. This article attempts to
examine the links between these events
and changes in the labour market over
the century, and in particular in the
changing nature of employment.
The changing population
Underlying the changes in the UK,
and the UK labour market, patterns over
the century have been changes in the
population, most notably in migration,
birth rates and life expectancy. Within
this, the main change has come from the
combination of falling birth rates and
reduced mortality, which have led to an
ageing population, with an increasing
proportion of people of post-retirement
age.
Migration
Britain is a country of immigration
and emigration. It has always been
Gross domestic product;a United Kingdom; 1900 to 2000
Million (£)
900,000
800,000
700,000
600,000
500,000
400,000
300,000
200,000
100,000
0
1900 1905 1910 1915 1920 1925 1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000
Source: United Kingdom National Accounts – The Blue Book
a At constant 1995 market prices. See technical note for further information about the data sources.
Note: Data have not been adjusted to reflect the post-2001 Census population estimates.
134 Labour Market trends
March 2003
Feature 5 March 03 (pp133-144)
27/2/03
5:09 pm
Page 135
A century of labour market change: 1900 to 2000
Figure
2
research,9 at the end of the century the
largest single identifiable group was UK
nationals (mostly returning emigrants,
although some were born abroad).
Other major sources were the European
Economic Area and Asia, but, overall,
there were significant numbers of
migrants from as many as 29 regions of
the world. This diversity of background
also feeds into labour market
experiences. Some migrants are very
successful, others are not. Immigrants
have a higher tendency to be
unemployed
or
particularly
economically inactive. Around 6 per
cent of migrants are unemployed
compared with a UK average of just
under 5 per cent. The migrant workingage employment rate is around 65 per
cent, compared with 74 per cent for the
UK as a whole. However, those who do
go into work, quite often move into
areas of labour shortage. For example,
according to a Home Office migration
report10 published in 2001, 31 per cent
of doctors and 13 per cent of nurses are
non-UK born; in London the figures are
23 per cent and 47 per cent respectively.
Half the expansion of the NHS over the
last decade – that is, 8,000 of the
additional 16,000 staff – had qualified
abroad. An estimated 70 per cent of
catering jobs in London are filled by
migrants. And in 1995-96, the Higher
Education Statistics Agency showed
that non-British nationals made up 12.5
per cent of academic and research staff,
were most likely to be in medicine,
science and engineering, and comprised
over half the faculty of the London
School of Economics.
Mortality
At the start of the twentieth century
the mortality rate was 16 deaths per
1,000 living persons,11 and had been in
constant decline since the 1860s when it
stood at around 22.5 per 1,000. As
Figure 2 shows, this sharp decline
continued into the first part of the new
century, with continuing falls up until
1921-25. The effect of the First World
War can be seen in the male mortality
series for 1916-20 when the rate
increased to 16.5 per 1,000. The rate
subsequently declined to 13 per 1,000
in 1921-25. Female mortality also
declined, reaching 11 deaths per 1,000.
By comparison, mortality rates have
been fairly flat in the period since 1925,
falling only marginally. Female
mortality was relatively unchanged at 11
deaths per 1,000 in 1991-95. The male
mortality rate had declined from 13 to 11
deaths per 1,000, excepting an upward
surge during the Second World War.
Most of this improvement in male
mortality has come after 1980. Looking
at the data on ‘deaths from injuries and
poisoning’,12 the main reason appears to
Mortality ratesa over five-year periods by sex; Great Britain; 1901-1905 to 1991-1995
Number of deaths per 1,000 people
18
16
14
12
10
8
6
All
4
Men
2
Women
0
1901-1905
1911-1915
1921-1925
1931-1935
1941-1945
1951-1955
1961-1965
1971-1975
1981-1985
1991-1995
Source: Office for National Statistics, Mortality Statistics, general 1998
a See technical note for further information about the data sources.
Note: Data have not been adjusted to reflect the post-2001 Census population estimates.
March 2003
Labour Market trends 135
A century of labour market change: 1900 to 2000
relatively open, and the British
population is now, as it always has been,
the result of successive influxes of
migrants and the racial and cultural
intermixture of these migrants with
those who were already here. It is also
reasonably clear, if difficult to quantify,
that Britain has benefited considerably,
in both economic and cultural terms,
from this openness. Patterns in
migration varied over the twentieth
century. The century began with net
emigration, mostly of young men, to
other countries, especially North
America, Australia, New Zealand and
the other colonies. During the Great
Depression this was reversed as
migrants returned home, and following
the Second World War there
was marked immigration from
Commonwealth countries. As a result,
there was net inward migration from
1931 through to 1961. In the 1970s and
1980s, the UK became a net exporter of
people as families migrated to Australia,
New Zealand and South Africa.
However, of late there has been a return
to net immigration. For example, in
2000, an estimated 100,000 more
people migrated into the UK than out of
it.8
Moreover, immigrants are diverse,
and cannot be easily treated as a single
group. Their backgrounds are diverse.
For example, according to Home Office
Special feature
Feature 5 March 03 (pp133-144)
A century of labour market change: 1900 to 2000
Special feature
Figure
3
27/2/03
5:09 pm
Page 136
A century of labour market change: 1900 to 2000
Number of people in employment;a United Kingdom; 1900 to 2000
Thousands
30,000
25,000
20,000
15,000
10,000
5,000
0
1900 1905 1910 1915 1920 1925 1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000
Sources: The Economist Publications Ltd.; Labour Force Survey
a See technical note for further information about the data sources.
Note: Data have not been adjusted to reflect the post-2001 Census population estimates.
have been the decline in deaths in road
traffic accidents (possibly linked to the
introduction of compulsory seatbelt
legislation in 1983).
Births
The number of births has declined
throughout the century, interrupted only
by the two post-war baby booms and a
peak in the 1960s. In 1900 the fertility
rate per woman was 3.5 children; by
1997 this had fallen to 1.7.
Linked to this, there has also been an
increasing tendency for women to have
their first child later in life. As a result,
the number of children born to women
aged under 30 has been in decline since
1961, whereas the number born to the
over 30s has been increasing.13
The changing nature of
employment and
unemployment
Given the growth in the UK population
it is not surprising that the most obvious
changes in employment have come in the
changes in the level of employment. In
1900 18 million people were employed in
the UK. By 2000 this had risen to 27
million (see Figure 3), mostly as a result
of population growth. As already noted,
the population increased from 38 million
to around 59 million.
136 Labour Market trends
March 2003
The employment level increased by
around 0.5 per cent per year on average.
However, within this, there have been
variations, most notably half a dozen
occasions when employment fell by 2
per cent or more. The most marked of
these was in 1921 when employment
fell 17 per cent at the onset of the interwar depression. The second largest fall,
of 5 per cent, came in 1946 following
the end of the Second World War. By
comparison, the largest falls in GDP
came in 1919-21 and 1944-45.
Similarly, it can be seen that
employment falls in 1981 and 1991-92
also appear to be lagged responses to
earlier falls in GDP. It is also noticeable
that the two periods with the most
marked falls in GDP both followed
world wars, which also provided the
periods of strongest GDP growth (17
per cent annual growth in 1915 and 14
per cent in 1940).
In terms of the employment rate, the
variations have been as marked as those
in the level, but the overall growth has
been somewhat less. It is possible to
calculate an estimate for the
employment rate for 15 to 64-yearolds14 in 1902 of around 69 per cent. On
the same basis, but using Labour
Force Survey (LFS) data, the current
employment rate in 2000 was 71 per
cent. However, this comparison is
affected by changes in the school
leaving age. The school leaving age was
12 years in 1900. This was raised to 14
in 1918 and then to 15 in 1947. This was
then raised again from 15 to 16 in
1972/73. Other things being equal, this
will have reduced the number of young
people in employment. For example,
according to the 1901 Census around 10
per cent (140,000) of 10 to 14-year-old
boys in Great Britain were already
‘engaged in occupations’. Meanwhile,
at the other end of the age scale, it is
important to be aware of the increase in
retirement; in 1901 nearly 40 per cent
(110,000) of men aged 75 or over were
still working. By comparison, in 2000
less than one in ten men was still in
employment after reaching the state
pension age of 65.
Just as employment has responded to
changes in output, so too has
unemployment. For much of the first 20
years of the twentieth century the
unemployment rate, as measured by
those claiming unemployment-related
benefit, was below 5 per cent (see
Figure 4). With the inter-war
depression, the rate increased to 17 per
cent in the early 1920s before easing to
around 10 per cent, and then soaring to
22 per cent in 1932. This fell back in the
run up to the Second World War,
dipping to below 1 per cent during the
war years. The post-war period was then
fairly stable, with unemployment below
Feature 5 March 03 (pp133-144)
27/2/03
5:09 pm
Page 137
A century of labour market change: 1900 to 2000
4
A century of labour market change: 1900 to 2000
Figure
Special feature
Year-on-year growth of the retail prices index (RPI)a and unemployment;b United Kingdom; 1900 to 2000
Per cent
30
25
20
15
10
5
0
-5
-10
-15
RPI
Unemployment rate
-20
-25
1900
1905
1910
1915
1920
1925
1930
1935
1940
1945
1950
1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
Sources: Economica; Retail Prices Index; Employment Gazette, December 1993; Office for National Statistics, claimant count series
a The RPI started in 1947. All figures before this date are estimated and are not official figures. They come from Layton and Crowther (1938). See technical note for further information on data sources.
b Unemployment data are based on register and claimant count data (see technical note for further information).
Note: Data have not been adjusted to reflect the post-2001 Census population estimates.
3 per cent until the mid-1970s when it
started to increase. The claimant
unemployment rate peaked in 1986 at
around 10.5 per cent and it was only in
the last few years of the century that it
dipped back to below 5 per cent.
The other interesting factor is the
changing
relationship
between
unemployment and inflation. The
standard Phillips Curve theory (see
technical note) suggests that there is a
trade-off
between
inflation
and
unemployment.
Low
or
falling
unemployment will, other things being
equal, increase inflation; high or rising
unemployment will lead to inflation
easing. Figure 4 shows that while this
holds for some periods, such as the 1920s
and 1930s, and maybe the 1960s or the
1980s, it is far from convincing over the
century as a whole. The most striking
contradictions of the rule are the 1970s,
when both inflation and unemployment
were rising, and the 1990s when both fell.
Industrial composition
As already mentioned, one major
change over the last century was the
shift in industrial composition, with
the decline of agriculture and
manufacturing’s
share
of
total
employment and the rise of services. In
the UK, manufacturing’s share fell from
28 to 14 per cent of employment, and
agriculture’s share from 11 to 2 per cent.
Nor was this a purely UK phenomenon.
By 1900, the UK had already undergone
the Industrial Revolution, and the
proportion
of
employment
in
agriculture was already in decline. By
comparison, according to Mitchell
(1998) agriculture still represented
around 40 per cent of employment in
France, Germany and the USA.15 By
1990, this was down to 6 per cent, 3 per
cent and 3 per cent respectively.
Similarly, all three countries had seen
growth in their service sectors, which
increased from 17 to 33 per cent of
employment in France, from 11 to 33
per cent in Germany, and from 14 to 33
per cent in the USA. On the same basis,
the UK saw service sector employment
increase from 21 to 32 per cent. This
differs from the 75 per cent quoted
earlier in the article, which is based on
official workforce jobs data and which
includes a wider number of industries
than are included in Mitchell’s estimate.
Mitchell excludes the transport and
communications, and commerce and
finance sectors from services. Including
these sectors in Mitchell’s estimates
would suggest that services have
increased from 34 per cent of UK
employment in 1901 to 70 per cent in
1991.
There were several drivers behind
this change, some local, some general.
Most importantly, there is nothing new
in shifting industrial composition;
before the Industrial Revolution most
people in the UK had worked in
agricultural industries. The eighteenth
and nineteenth centuries saw that
change as technological innovations
began to improve the productivity of
agricultural workers, starting with
Jethro Tull’s (1674-1741) mechanical
seed
sower
in
1701.
These
developments meant that more
agricultural produce could be provided
by fewer workers and consequently the
numbers employed in the primary
March 2003
Labour Market trends 137
Feature 5 March 03 (pp133-144)
A century of labour market change: 1900 to 2000
Special feature
27/2/03
5:09 pm
Page 138
A century of labour market change: 1900 to 2000
sector began to fall. However, at the
same time, technology such as the steam
engine
triggered
the
Industrial
Revolution. The decline in employment
in agriculture was more than offset by
the increase in the new manufacturing
and mining sectors. Similarly, the
twentieth century saw ongoing
improvements in technology which
improved
efficiency
in
the
manufacturing sector, for example
computers and automation. This in turn
freed up resources to work in the
burgeoning services sector. Moreover,
each revolution also helped fuel the
next: improved efficiency in agriculture
raised living standards and wealth
which increased the demand for
manufactured goods; similarly, rising
living standards and leisure in the
twentieth century helped increase the
demand for services.
Alongside this is a more international
aspect to the process. Another driver
behind manufacturing’s decline was
competition from abroad. To take one
example, 1913 was a record year for the
Lancashire cotton industry. Exports of
woven cloth from the region topped
7,000 million linear yards – around 65
per cent of world output.16 By 1960, the
Lancashire cotton industry was dead,
killed by a combination of lack of
investment, which left it at a
disadvantage when competing with
newcomers
with
newer,
better
machinery, and the fact that overseas
labour was cheaper. The pattern was to
be repeated across a number of
industries, and the economy moved to
focus on those areas where it had more
of a competitive advantage, such as
services or those manufacturing
industries that required higher skills.
Participation in the
workforce
Another major long-term change has
been the increasing female participation
in the workforce. At the beginning of
the twentieth century, around five
million women worked, making up 29
per cent of the total workforce.17 By
2000 the figure had risen to 13 million,
representing around 53 per cent of the
female population aged 16 and over and
46 per cent of the total workforce.
The First World War provided an initial
138 Labour Market trends
March 2003
opportunity for women to move into
industry, with men fighting overseas.
However, longer-term growth in female
participation is linked to other changes,
some within the labour market and some
within society more generally. The more
general rise in female emancipation and
the change in attitudes around the 1960s
are likely to have helped, and it is
noticeable that the major increase in
female participation has come in the postwar period. However, there are also other
direct economic factors at play. The first
is the general move, already noted, from
heavy industry to services. Even within
remaining manufacturing there has been
a shift from old industries to automation
and to hi-tech manufacture, which are
less physically demanding and,
consequently,
potentially
more
accessible. For example, the move from
shipbuilding to computer component
manufacture.
Also, within the labour market, there
has been the increased use of part-time
workers,
leading
to
ongoing
developments in flexible working. The
development of part-time working was
aided by the rise of the service sector and
in turn made it easier for women with
families to return to work. This move
was probably also aided by the increased
use of labour-saving equipment in the
home, for example washing machines
and tumble dryers among others.
Outside of the labour market, another
development which helped increase
female participation was the rise of
widespread education. Government
involvement in education before 1900
had been limited. The 1870 Elementary
Education Act had required local school
boards to provide elementary schools
where
existing
facilities
were
inadequate, but it was not until the 1902
Education Act that schooling came
under local authority control. It was also
not until 1902 that the Government
made any effort to establish a system of
secondary education, when the Act
provided for two types of state-aided
secondary school: the endowed
grammar school; and the municipal or
county secondary schools. It was the
1944 Education Act that then
established the principle of free
education for all from primary to
secondary level. These changes in
education seem likely to have fuelled
increased female participation in two
ways: first, with their children at school,
women were more able to take up
employment; secondly, and probably
more importantly, with increased
education women were equipped to take
on the new jobs. For example, in 1922
female students obtained around 23 per
cent of all first degrees – out of a total of
just over 10,000; by 1993, this had risen
to 45 per cent of around 90,000 degrees
awarded in the UK.18
By comparison, male participation
rates in the labour force fell over the
latter part of the twentieth century. In
part, this seems to have been associated
with the same industrial shift which has
helped increase female participation.
For example, the decline of old heavy
industries such as coal, shipbuilding and
steel left a large number of men
unemployed in the 1980s. Many seem to
have drifted into inactivity, feeling
detached from the labour market –
either too old or unwilling to reskill. It
might be expected that this is to be a
passing problem that will lessen as the
particular affected cohort of workers
ages and leaves the workforce.
However, there does appear to be a
more persistent and general decline in
male economic activity.
This increase in male inactivity
appears to be associated with an
increase in the ill health of the inactive
towards the end of the century. In
particular, since 1981 there have been
increasing numbers of working-age
people being reported as disabled or
long-term sick. For example, in 199798, almost two million long-term sick
and disabled people in Great Britain
were in receipt of incapacity benefit or
severe disablement benefit, which was
more than double the number on the
equivalent benefits in 1981-82. The
number of people receiving disability
living allowance was three million, up
over five times. Not surprisingly, this
has fed into the social security budget,
with social security expenditure more
than doubling in real terms between
1977-78 and 1999-00 to stand at almost
£103 billion.
The question is: why has there been
this increase in ill health? The rise in the
number of those receiving these
Feature 5 March 03 (pp133-144)
27/2/03
5:09 pm
Page 139
A century of labour market change: 1900 to 2000
5
Proportion of people in employment who are members of a trade union; United Kingdom;a 1900 to 2000
Per cent
60
50
40
30
20
10
0
1900
1905
1910
1915
1920
1925
1930
1935
1940
1945
1950
1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
Sources: Trade union membership levels: British Labour Statistics Historical Abstract 1886-1968; Department of Employment;
Certification Officer's Annual Reports. Employment levels: One Hundred Years of Economic Statistics; Labour Force Survey
a Data since 1975 from the Certification Officer's Annual Reports are for Great Britain only. See technical note for further information about the data sources.
Note: Data have not been adjusted to reflect the post-2001 Census population estimates.
Figure
6
Number of working days lost due to labour disputes;a United Kingdom; 1901 to 2000
Millions
180
160
140
120
100
80
60
40
20
0
1901
1906
1911
1916
1921
1926
1931
1936
1941
1946
1951
1956
1961
1966
1971
1976
1981
1986
1991
1996
2000
Source: Social Trends 30, Office for National Statistics
a See technical note for further information about the data sources.
Note: Data have not been adjusted to reflect the post-2001 Census population estimates.
invalidity benefits is partly due to an
increase in the duration of claims rather
than an increase in new claims.
However, as already noted there has
also been a real increase, particularly
among men, which in part seems to be
linked to the decline of certain
traditional industries and the resulting
impacts on local areas. The rise cannot
be explained solely in terms of the
cohort of workers who left the old
nationalised industries as subsidies
were cut; the problems continue to
affect the next generation, and the
explanation appears to be at least in part
cultural, with a cohort of workers
growing up used to worklessness. For
example,
a
regional
strengths,
weaknesses, opportunities and threats
(SWOT) analysis (see technical note)
carried out in East Wales concluded that
certain areas were affected by the
decline and restructuring of traditional
industries. People experienced multiple
deprivation which ‘. . . contributes to an
inter-generational cycle of inactivity,
low expectations, poor skills acquisition
and social exclusion’.19 Alongside this is
an increased awareness of illness and
disease, such as depression. Previously
such disorders may have existed but
simply gone undiagnosed.
Hours worked
Alongside
the
changes
in
employment type and characteristics
March 2003
Labour Market trends 139
A century of labour market change: 1900 to 2000
Figure
Special feature
Feature 5 March 03 (pp133-144)
A century of labour market change: 1900 to 2000
Special feature
27/2/03
5:09 pm
A century of labour market change: 1900 to 2000
there was also a general decline in
average hours worked over the
twentieth century. In 1870 annual hours
worked per person stood at 2,984. By
1913 this was down to 2,624 and the
decline continued, reaching 1,489 in
1998.20 Similar trends can be seen across
the developed world, and are linked to
technological
change
increasing
productivity. This rising productivity in
turn feeds into rising wages, and as
wages increase beyond the subsistence
level the greater the demand, and
opportunity, for increased leisure time.
The decline in annual hours can also
be seen in the reduced length of the
average working week. For example,
Figure
Page 140
the average weekly hours of a manual
worker fell from 53 hours in 1943 to
43.5 in 1987.21 Moreover, while overall
hours have fallen there have been
changes in working patterns, which
have altered the nature of the working
week. For example, Sunday working
has become more widespread since the
Sunday Trading Act 1994, which
allowed Sunday shop opening in
England and Wales.
employment. This rose to 40 per cent in
the early 1920s before falling back to
around 24 per cent in the mid-1930s
(see Figure 5). Membership then surged
again, reaching 40 per cent in the late
1940s and remained fairly constant until
the 1970s when again recruitment
increased. Union membership peaked in
the late 1970s at a little over 50 per cent
of those in employment. However, it has
been in almost continuous decline ever
since and by 2000 was down to under 29
per cent, its lowest level for 60 years.
Coupled to this there were varying
levels of industrial unrest through the
century. As Figure 6 shows, for much of
the period, the number of working days
Workplace relations
The century also saw major changes
in workplace relations. In 1900 trade
union membership represented a little
under 11 per cent of those in
Index of earnings;a United Kingdom; 1900 to 2000
7
1987=100
250
b
The index of Liesner's
hourly earnings
c
Average Earnings Index
200
150
100
50
0
1900
1905
1910
1915
1920
1925
1930
1935
1940
1945
1950
1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
Sources: The Economist Publications Ltd.; Average Earnings Index
a See technical note for further information about the data sources.
b The Liesner index data are based on the hourly earnings rates given in Liesner's One Hundred Years of Economic Statistics. Data have been
converted into an index so that 1987=100.
c The Average Earnings Index is based on the published data. However, the index has been reweighted so that 1987=100 instead of 1995=100
as in the published data.
Note: Data have not been adjusted to reflect the post-2001 Census population estimates.
Figure
8
Growth rates of average earnings and the Retail Prices Index; Great Britain; 1940 to 2000
Per cent
30
The index of Liesner's
hourly earnings
Average Earnings Index
Retail Prices Index
25
20
15
10
5
0
-5
1940
1945
1950
1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
Sources: The Economist Publications Ltd.; Average Earnings Index
140 Labour Market trends
March 2003
Feature 5 March 03 (pp133-144)
27/2/03
5:09 pm
Page 141
A century of labour market change: 1900 to 2000
9
A century of labour market change: 1900 to 2000
Figure
Special feature
Proportions of households with regular access to a car;a United Kingdom; 1951 to 2000
Per cent
100
No car
90
One car
80
Two or more cars
70
60
50
40
30
20
10
0
1951
1956
1961
1965
1971
1976
1981
1986
1991
1996
2000
Sources: Family Expenditure Survey; General Household Survey; National Travel Survey
a See technical note for futher information about data sources.
lost to stoppages has remained fairly
low. The first part of the century saw
three main periods of unrest, each more
marked than its predecessor. This
culminated in the General Strike in
1926 when 162 million days were lost to
strikes; this was more than was lost for
the entire period between 1927 and
1970, an era of relative industrial peace.
The 1970s and 1980s then saw
industrial action flaring up again with
three years in which days lost hit 20
million. However, following the end of
the miners’ strike in 1985 and reform of
union legislation, the number of
stoppages fell away again. It is
important to note that a single major
stoppage can dominate these figures.
For example, even of the 162 million
days lost to strikes in 1926, 90 per cent
were in the coal industry.
Earnings
The level of full-time earnings has
soared from an average £1.40 per week
in 1902 (unadjusted for inflation) to
£350 per week in 1997. Figure 7
illustrates an index of weekly earnings
over the century with 1987=100. As can
be clearly seen most of this increase
came in the period from 1970 onwards.
However, this is slightly misleading.
The patterns in growth are more
accurately drawn out in Figure 8, which
provides annual growth rates for
earnings and the retail prices index
(RPI). Most significant is the peak in the
1970s, which is also in official average
earnings index (AEI) data, and which
can be linked to the high inflation of the
period, and the resulting wage-price
spirals. Generally, and without
broaching the issue of causality,
earnings and inflation have moved
together since the 1950s.
Figure 8 also brings out, which is not
clear in the previous figure, that wage
rate inflation has been declining since
the 1970s.
Labour mobility
Linked to the changing nature of
work and the increased mobility of the
workforce has been the rise of motor
transport, in particular the car. The car
has revolutionised working patterns,
increasing travel and labour mobility,
and allowing an increase in distances
commuted. Data on the first half of the
century are not available, but even in
1951 only 14 per cent of households had
regular access to a car (see Figure 9). By
2000 this had increased to 73 per cent.
The big growth came between 1951 and
1970 with the proportion of households
with access to one car rising from 13 to
45 per cent by 1970. This has remained
constant since, but there has been
continuing growth in the proportion of
households with two or more cars (from
7 per cent in 1970 to 28 per cent in
2000).
As well as the increase in car
ownership, there has been an increase in
commuting. For example, from 1976 to
1999/2001 average commuting trip
distance increased by over 60 per cent.22
There are various reasons for increased
commuting in recent times. Greater
specialisation in the job market may
have led to more distant opportunities
and more frequent job moves. The
growth in female participation has
increased the number of households in
which both partners are working, and
inevitably in some cases travelling in
opposite directions and living in the
middle. Some commuting is from
choice. Nor is all commuting by road.
Travel has also became easier with the
development of the railway, and for
certain parts of the country – in
particular London – there can be no
doubt that the development, and spread,
of commuting is heavily linked to the
development of the railway.
The key point is that over the last
century travelling to work has become
easier and allowed people to commute
March 2003
Labour Market trends 141
Feature 5 March 03 (pp133-144)
A century of labour market change: 1900 to 2000
Special feature
27/2/03
5:09 pm
Page 142
A century of labour market change: 1900 to 2000
longer distances. As such, it has had the
effect of increasing the catchment area
of local labour markets (as defined by
Travel-to-Work Areas).
Conclusion
The twentieth century was a period of
great change. In some cases, these were
trends that had started in the previous
century, for example in terms of
industrial change and the continuing
improvements in technology. As always
when there is such sweeping change,
there are winners and losers. Many
traditional
industries
such
as
shipbuilding or mining, growth areas of
the nineteenth century, went into
decline. Trade unions rose and then fell
in influence. However, the overall
improvements were overwhelming:
better working conditions generally;
falling hours; increased real wages;
greater flexibility in work; and
increased female participation in the
workforce.
i
Further information
For further information, contact:
Craig Lindsay,
B3/04,
Office for National Statistics,
1 Drummond Gate,
London SW1V 2QQ,
e-mail [email protected],
tel. 020 7533 5896.
Notes
1
House of Commons Library Research Paper 99/111, A Century of Change:Trends in UK Statistics since 1900 (1999).
2
100 Years of GDP 1900-1999 at: http:/statbase/themes/economy/Articles/NationalAccounts/Articles/100_years_of_GDP.asp.
3
Mitchell, B. R., International Historical Statistics: Europe 1750-1993, fourth edition.
4
2001 Census of Population.
5
House of Commons, Research Paper 02/44, Inflation: the value of the pound 1750-2001 (2002) at
6
Workforce jobs.
7
Instititue for Fiscal Studies Briefing Note 25: Long-Term Trends in British Taxation and Spending (2002).
8
Office for National Statistics, International Migration 2001– interim estimates, First Release (28 November 2002).
9
Home Office, Research Study, The Settlement of Refugees in Britain, No.141 (1995).
10
Home Office, Research Study, RDS Occasional Paper No. 67, Migration: an economic and social analysis, at
www.parliament.uk/commons/lib/research/rp2002/rp02-044.pdf.
www.homeoffice.gov.uk/rds/pdfs/occ67-migration.pdf.
11
Office for National Statistics, Mortality Statistics general 1998 (series DH1, No. 31), at www.statistics.gov.uk/statbase/xsdataset.asp?vlnk=2278.
12
Office for National Statistics,‘Deaths from injury and poisoning: external cause and year of registration or occurrence, 1901-2000’ taken from Table 3
published in Mortality Statistics: Injury and Poisoning (series DH4, No. 25), at www.statistics.gov.uk/statbase/xsdataset.asp?vlnk=5679.
13
Table 3.1 of Population Trends, No. 110, at www.statistics.gov.uk/statbase/xsdataset.asp?vlnk=6161.
14
Based on population and employment data in Liesner,T. Publications Ltd, One Hundred Years of Economic Statistics,The Economist Publications Ltd. (1989).
15
Mitchell, B. R., International Historical Statistics: Europe 1750-1993, fourth edition, and Mitchell, B. R., International Historical Statistics:The Americas 1750-1993,
16
History and Heritage of the City and the Metropolitan County of Greater Manchester, at
fourth edition.
www.manchester2002-uk.com/history/modern/20thcent-1.html.
17
Mitchell, B. R., International Historical Statistics: Europe 1750-1993, fourth edition.
18
House of Commons Library Research Paper 99/111, A Century of Change:Trends in UK Statistics since 1900 (1999).
19
East Wales, Objective 3, Regional Action Plan, Appendix 5, at prp.powys.org.uk/docs/obj3%20rap%20appendix%205.pdf.
20
Maddison, A., The World Economy: A Millennial Perspective, Organization for Economic Cooperation and Development (2001).
21
Liesner,T. Publications Ltd, One Hundred Years of Economic Statistics,The Economist Publications Ltd. (1989).
22
Department for Transport, National Travel Survey 1999/2001. Update at www.transtat.dft.gov.uk/tables/2002/nts/nts02.htm.
142 Labour Market trends
March 2003
Feature 5 March 03 (pp133-144)
27/2/03
5:09 pm
Page 143
A century of labour market change: 1900 to 2000
Special feature
Allin, P.,‘One Hundred Years of Labour Market Indicators’, Employment Gazette, pp553-7, December 1993.
Department of Employment and Productivity, British Labour Statistics Historical Abstract 1886-1968 (1971).
Phillips, H.,‘The Relation between Unemployment and the Rate of Change of Money Wage Rates in the United Kingdom’, 1861-1957, Economica, new Series 25,
No.2, (1958).
Home Office Research Study,‘The Settlement of Refugees in Britain’, RDS Occasional Paper No. 67, Migration: An economic and social analysis, No.141 (1995).
House of Commons Library Research Paper 99/111, A Century of Change:Trends in UK Statistics since 1900 (1999).
House of Commons Library Research Paper 02/44, Inflation: the value of the pound 1750-2001 (2002).
Institute for Fiscal Studies Briefing Note 25: Long-Term Trends in British Taxation and Spending (2002).
Layston,W.T. and Crowther, G., An Introduction to the Study of Prices, Macmillan and Co., (1938).
Liesner,T. Publications Ltd, One Hundred Years of Economic Statistics,The Economist Publications Ltd. (1989).
Maddison,A., The World Economy: A Millennial Perspective, Organization for Economic Cooperation and Development (2001).
Mitchell, B. R., International Historical Statistics: Europe 1750-1993, fourth edition, New York Stockton Press (1998).
Mitchell, B. R., International Historical Statistics:The Americas 1750-1993, fourth edition, New York Stockton Press (1998).
Technical note
This article presents an overview of the main trends within
the UK labour market over the twentieth century. In doing this,
it uses data from a number of different sources, and inevitably
there are issues of consistency and comparability over time. Such
issues of consistency occur throughout this analysis. Data have
only been presented where it is considered that they add value.
However, this is an analytical article trying to draw out the main
messages from the data and not a reconciliation piece; users
should be aware that there are limitations in comparing data over
such a long period of time.
Liesner, published in cooperation with ONS in 1989. Data for
later years are consistent with the 2000 edition of The Blue Book.
Data for years before 1948 are not available from National
Statistics.
Mortality rates
Figure 2 shows mortality rate data which are taken from
Table 3 published in Mortality Statistics, general 1998 (series
DH1 No. 31), Appendix 1 (notes to tables – sources,
methods
and
definitions)
and
are
available
at
www.statistics.gov.uk/statbase/xsdataset.asp?vlnk=2278.
Chart sources
Gross domestic product (GDP)
Employment
Figure 1 showing historical GDP is reproduced from the
article ‘100 Years of GDP 1900-1999’, which was originally
published in the 2000 edition of the United Kingdom National
Accounts – The Blue Book, and is available online at
statbase/themes/economy/Articles/NationalAccounts/Articles/1
00_years_of_GDP.asp.
The source for much of the data before 1948 is The Economist
publication, One Hundred Years of Economic Statistics by Thelma
Employment data are taken from two main sources: One
Hundred Years of Economic Statistics for the period 1900-1983; and
the Labour Force Survey from 1984-2000. This does create a
potential inconsistency as the former is based on Census of
Employment data, whereas the latter is based on the number of
people in employment. However, this does not detract from the
overall message seen in the data, and the two series do appear
reasonably compatible.
March 2003
Labour Market trends 143
A century of labour market change: 1900 to 2000
References
Feature 5 March 03 (pp133-144)
A century of labour market change: 1900 to 2000
Special feature
27/2/03
5:09 pm
Page 144
A century of labour market change: 1900 to 2000
Technical note
Historical industrial employment data are derived from B. R.
Mitchell’s International Historical Statistics: Europe 1750-1993,
fourth edition.This allows comparison with other countries on a
more comparable basis. However, One Hundred Years of Economic
Statistics does contain some industrial data, and while the detail
differs, the overall message is the same: a decline in agriculture
(from 13 to 1 per cent of the workforce); and manufacturing
(from 33 to 20 per cent) between 1901 and 1987. The main
difference is in services, where the sector is more tightly defined,
and so smaller both in 1901 and 1987. But again, it has seen
growth (from 20 to 35 per cent).The other sectors identified by
One Hundred Years of Economic Statistics are largely unchanged:
construction, energy and water supply, transport and
communication have all declined slightly as a proportion of the
workforce (by 2 to 4 percentage points). Distributive trades and
public administration and defence have increased (by 1 to 2
percentage points).
Unemployment
Data on unemployment are based on the number of
registered unemployed in the British Labour Historical Abstract
1886-1968 for the period 1900 to 1968. Data for 1969-79 are
taken from Employment and Productivity Gazette, vol. LXVIII,
January-December 1970. Data for the period 1971 onwards are
taken from the claimant count series available on the National
Statistics website. This has a number of consistency issues. The
series from 1971 to the present has been adjusted to be on a
consistent basis, but before that the data are affected by different
definitions as set out in the Historical Abstract. Most notably, up
until 1948 the series is based on the proportion of insured
workers unemployed – but not all workers were covered by the
scope of the Unemployment Insurance Acts. For example, certain
domestic employment was only brought within scope in 1938.
From 1948 onwards the data are based on the unemployment
register.
In addition, while the claimant series is the only long-term
series available, the user should be aware that it is a narrower
measure and does not correspond to ‘unemployment’ as defined
under National Statistics today.
Retail prices index (RPI)
Figure 4 shows retail prices growth, which has been calculated
from the RPI. However, the RPI does not date back to 1900.
Official RPI data started in 1947. All figures before 1947 are
estimated and are not ‘official’ figures, and come from estimates
in Layton and Crowther, An Introduction to the Study of Prices
(1938).
Trade union membership
The data on trade union membership levels are drawn from
three different sources. For the period 1900-1968 figures come
from the British Labour Statistics Historical Abstract 1886-1968 and
are on a UK basis. From 1968-74, the figures are from the former
Department of Employment, and are also on a UK basis. Data
from 1975 are taken from Certification Officer’s Annual Reports
and are for Great Britain (GB).Trade union membership rates are
then calculated as a proportion of the employment series
outlined earlier.The switch from UK to GB does mean that there
is an inconsistency in the series, but the impact appears to be
small and does not detract from the overall message of union
membership.
144 Labour Market trends
March 2003
Working days lost through stoppages
The historic figures for working days lost are taken from
Social Trends 30 and are available on the National Statistics website at
www.statistics.gov.uk/statbase/xsdataset.asp?vlnk=134&more=Y.
Earnings
Two main sources are used for earnings data: The Economist
publication One Hundred Years of Economic Statistics; and the
Average Earnings Index (AEI). The former provides data on the
average weekly earnings of manual workers. This has been
converted into an index with 1987 equal to 100 for the purposes
of Figure 7. The AEI is available from 1963 – although not on a
completely consistent basis.The AEI is normally presented as an
index with a base year of 1995. For the purposes of Figure 7, it
was rebased to 1987=100. The reason for this was to allow a
better comparison with the data from the publication One
Hundred Years of Economic Statistics; 1987 was the last year of
available data from The Economist publication.
Car ownership
Data on car ownership is taken from Transport Statistics Great
Britain: 2002 Edition. Figure 7 is derived from a number of surveys:
the National Travel Survey, the Family Expenditure Survey and
the General Household Survey – and is available at
www.transtat.dft.gov.uk/tables/tsgb02/9/section9.htm#9.04.
‘Phillips Curve’
The Phillips Curve represents the relationship between the
rate of inflation and the unemployment rate, and was outlined by
A. W. H. Phillips in his 1958 study ‘The Relation between
Unemployment and the Rate of Change of Money Wage Rates in
the United Kingdom, 1861-1957’ (Economica, NS 25, No. 2, 1958).
Phillips discovered that there was a consistent inverse, or
negative, relationship between the rate of wage inflation and the
rate of unemployment in the UK from 1861 to 1957. When
unemployment was high, wages increased slowly; when
unemployment was low, wages rose rapidly. (The only important
exception was during the period of volatile inflation between the
two world wars.) This in turn led to the suggestion that there is a
trade-off between unemployment and inflation. However, this
inverse relationship did not hold throughout the twentieth
century, most notably breaking down in the 1970s when both
inflation and unemployment reached high levels. For more
information,see www.econlib.org/library/Enc/PhillipsCurve.html.
SWOT analysis
SWOT analysis is a business tool for auditing an organisation
and its operations, for informing planning and helping a company
to focus on key issues. SWOT stands for ‘strengths, weaknesses,
opportunities, and threats’. Strengths and weaknesses are internal
factors. For example, a strength could be specialist expertise. A
weakness could be a high cost structure. Opportunities and
threats are external factors. An opportunity could be a new
technological development; a threat could be a shift in consumer
tastes. Successful businesses build on their strengths, correct
weaknesses and protect against vulnerabilities and threats. For
further information,see www.quickmba.com/strategy/swot/.